After some careful consideration and examination of the original Lawnbot design, I was able to make some decisions: First off, after pricing the components for the original design’s H-bridge board (including the supplies and equipment I would need to acquire in order to start etching my own PCB’s), I decided to go with an off-the-shelf solution, Dimension Engineering’s Sabertooth 2×25. The board has some serious advantages for my purposes. It’s small, it’s ready to go (okay, that’s not necessarily a big plus, since this is supposed to be an educational experience), it has overcurrent and overtemperature protections (Definitely a big plus!), and most importantly, it will accept standard RC inputs. As far as I can tell, the only reason the original Lawnbot used an Arduino board (at least at the most recent stage) was to translate the RC signal (PPM or Pulse Position Modulation) into Pulse Width Modulation (PWM). Since the Sabertooth does this by itself, I can remove the Arduino from the current configuration of the Robomower, thus keeping the simplicity high, and the potential sites of tragic errors to a minimum. I am definitely still excited about etching PCBs, but there are plenty of other projects out there which are better suited for first-time etching. So anyway, the Sabertooth has been ordered, and hopefully will be here shortly.
I also did some research, and decided to go with the Spektrum DX5e RC transmitter and the BR6000 receiver. The DX5e seems like a great choice for a robot: it has 5 channels (with a sneaky way to get a sixth!), it transmits at 2.4ghz, so theoretically, there’s a lot less trouble with things like chain-link fences, not to mention no channel confusion, so the kid across the street can’t accidentally go off on a Stephen King style robot rampage while wondering why his RC car isn’t doing what he wants it to do. It also has the serious advantage of being at a VERY attractive price point relative to the other similar offerings on the market.
The BR6000 also looks like a pretty obvious choice. It will receive 6 channels (and will allow the DX5e to use the sneaky one), and most importantly, it has a fail-safe on every single one. This is a very good thing. If the mower loses signal, I want both drive motors to stop right where they are, and I want the mower motor to do the exact same thing!
Well, I was able to find the DX5e locally, which was nice! I did have to special order the BR6000, but at least I’m not paying for shipping. On the way home, I discovered what is for me one small problem with the DX5e: the left-hand joystick only self-centers in the horizontal axis. That’s not really what I’m looking for, since the Robomower will use “Tank Steering”. My preference would be for the sticks to center in the vertical axis. The good news is that it took me about an hour to crack open the case and discover that the joystick case is actually molded to accept either a friction ratchet or a centering spring. So I just took the spring part off of the horizontal axis, and moved it to the vertical. Now, of course, the stick does not center horizontally, but I can live with that until I either find another spring part, or just go get a spring and cut the bracket out of sheet styrene. Either way, I am happy for the foreseeable future.
So, the motor controller is on order, as is the RC receiver, and the transmitter is in my hands. The next step: start building a frame for those motors!